Neural networks have the problems of excessive number of parameters and computation due to data noise and redundant filters, which limits the application of neural networks. A feature map pruning method using channel attention (FPC) is proposed to address the problem of excessive accuracy loss of existing pruning methods. Experiments show that the feature graph pruning method using channel attention has better pruning effect with the same accuracy loss.
In view of the low recognition accuracy of traditional weather recognition methods and the serious imbalance in the number of weather images in various categories in the weather image data set, a weather image classification algorithm based on generative adversarial network and transfer learning is proposed to solve the above problems. The proposed method mainly includes two parts: data equalization based on generative adversarial network and image classification based on transfer learning. This paper uses generative adversarial network to amplify the data of a few categories of weather images, so as to obtain a relatively balanced weather image data set.The method of transfer learning is used to fine-tune the model to realize the classification of weather images. The experimental results show that the method proposed in this paper is better than the traditional method, effectively solving the problem of low model classification accuracy caused by the imbalance of training samples, and realizing the recognition and classification of four types of weather images: sunny, foggy, rainy, and snowy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.